Providing graduated corrective feedback in an intelligent computer-assisted language learning environment

Corrective feedback (CF), a response to linguistic errors made by second language (L2) learners, has received extensive scholarly attention in second language acquisition. While much of the previous research in the field has focused on whether CF facilitates or impedes L2 development, few studies have examined the efficacy of gradually modifying the explicitness or specificity of CF as a function of a learner's response to the feedback. Yet, the type and extent of CF needed by a learner, as suggested by Vygotsky (1978), sheds light on whether a learner is developing his or her abilities in a particular area and the ways in which they do it. This paper reports on a study that explores the design, effectiveness and learners' perception toward a graduated (Aljaafreh & Lantolf, 1994) approach to CF, i.e., feedback that progresses from very general and implicit to very specific and explicit, in an intelligent computer-assisted language learning (ICALL) environment. The results show that the graduated approach to CF is effective in helping learners to self-identify and self-correct a number of grammatical issues, although an onsite tutor provides necessary remedies when the ICALL system occasionally fails to do its part. Implications for CF research, particularly on the notion of individualized feedback, are also discussed.